Richard Wagner fact-checks a recent article from the New York Post, claiming that Trump suddenly gained 16.5 percentage points among African-American voters in a recent LA Times poll.
Trump, like most Republican Presidential candidates since Ronald Reagan, has had about 5% or less of African American support. If a Republican candidate get above 10% of African-American support, it’s a big deal.
Therefore, this recent New York Post article should raise skepticism, as it claims the following:
Trump saw a 16.5 percentage-point increase in backing from African-American voters in a Los Angeles Times/University of Southern California tracking poll, up from 3.1 percent on Sept. 10 to 19.6 percent through Friday.
A google search will likely lead you to more questionable sources, like Breitbart and Newsmax, repeating the story, and their only source for this claim being the above mentioned New York Post article. However, the New York Post article contains no hyperlinks to this poll. It only claims that it is a poll from the “Los Angeles Times/University of Southern California”.
However, after some more extensive research, here is the breakdown of this poll. If you follow the link and scroll down about half way, you can see support “By Race/Ethnicity”. New York Post was only slightly off. Trump did get the support of 18.9% of black participants, less than 1% below the “19.6%” claimed by New York Post. This is hardly enough of an error to bring the New York Post’s credibility into question.
So yes, in this particular poll, Trump really did make a substantial gain among African Americans. If this actually continued all the way to Election Day, it would be historically monumental.
Keeping it in perspective
However, it’s just one poll. The sample size for these LA Times “daybreak” polls are a little small, at 400, but not so small that they can’t be taken seriously. (A standard public opinion poll has a sample size of 1000).
It’s possible that for one reason or another, about 15% of African Americans have suddenly decided to support Trump, which may or may not last until November. It’s also possible that this one poll was a fluke. If you randomly ask 400 people out of the roughly 300 million size US population, it’s likely that about 50 are African American (12%).
Hypothetically, let’s say that this poll is wrong and that only 3% of the black population of the US supports Trump. That would mean that an accurate poll with 50 black participants would only have about 2 black Trump supporters. As you can see, it would be very easy for a legitimate poll, but mere chance, to happen to pick a group of 50 African Americans that contains 9 Trump supporters, instead of 2, causing them to show that about 19% of blacks support Trump.
Time will tell if more polls in the near future show similar results, or if this LA Times poll is just a fluke.
Gino Santa Maria / Shutterstock.com